I found this condensed passage below very interesting. I think it should be read in conjunction with my latest article, My Investment Approach.
It should never be forgotten that, in its most basic form, investing is always and everywhere about price and value. Price is what you pay, says the Sage of Omaha, and value is what you get. By this definition, every serious investor must be a value investor. This is not to say that investors should restrict themselves to buying companies with low valuation multiples. The business of investment is ultimately about buying stocks at a discount to intrinsic value.
So how do you calculate value? Well, in theory the value received is derived from future cash flows discounted back to today at the appropriate discount rate. The trouble is that we are rather poor at making predictions, especially about the future. But that doesn’t put us off. We suffer from what Nassim Taleb calls the “epistemic arrogance” – in plain English, we think we are better at making predictions than we really are. The result is that we have a misplaced sense of confidence in our forecasts. Investors like modelling because it appears scientific (the more spreadsheet tabs, the greater the effect).
Investment models, however, encourage anchoring. Most models are calibrated to produce a current value for a company within a reasonable range of the current price. Another wrinkle is the discount rates. If you don’t accept that historical volatility (beta) is a good measure of risk (which we do not), then it’s not clear how to calculate the appropriate discount rate. At Marathon, we believe that detailed forecasting adds little value.
One common response to the difficulty of forecasting is to turn to simple value proxies, such as the price-to-book ratio, price-to-earnings ratio, and free cash flow yield. Many “value” investors advocate buying a basket of stocks which are cheap by these measures. There’s nothing inherently dumb about this approach. Each of the measures is a very useful indicator of potential value, but there’s a danger of oversimplification. Traditional valuation measures say nothing about the specific context of an investment – for instance, a company’s business model, its industry structure, and management’s ability to allocate capital – which determines future cash flows.
Quantitative valuation measures also tend to encourage a narrow categorization of investment styles. Take for example the S&P US Style Indices. Value stocks are defined by their ratios of price-to-book, price-to-earnings, and price-to-sales. The growth index, on the other hand, is defined by the three-year change in earnings per share, three-year sales per share growth rate, and 12-month price momentum. While some of these factors are powerful, they are too crude to be the sole framework for assessing value.